AI/ML Engineer (Applied)

Role overview

This is an applied AI/ML engineering role focused on building and shipping LLM-powered features inside a production voice, video, and documentation platform. The work spans model selection, evaluation design, and tight integration with product code, with a clear emphasis on pragmatic shipping rather than frontier research. Expect close collaboration with founders and the ability to land features in production within days rather than quarters.

Responsibilities

- Design and ship LLM-backed product features end to end, from prototype through production - Build and maintain evaluation frameworks, designed rubrics, or automated LLM-as-judge pipelines that measure quality in live systems - Integrate models with the product codebase, including pipelines that support voice, transcription, scripting, and documentation workflows across many languages and accents - Choose the smallest model that solves the problem, then instrument it for reliability, latency, and cost - Partner with leadership to scope experiments, define quality bars, and turn research-style results into reliable shipped product

Requirements

- Hands-on experience shipping LLM-powered systems in production - Comfort with Python plus at least one typed language (TypeScript, Go, or Rust) for integrating with product code - Practical experience with LLM evaluations, designed rubrics, or LLM-as-judge pipelines running at scale - Pragmatic shipping mindset, favoring simple, effective solutions over fashionable complexity - Strong written communication suited to a remote, async-first team

Nice to have

- Experience with fine-tuning, LoRA, DPO, or related adaptation techniques - Background in speech (Whisper, TTS, voice cloning) or other multimodal models - Familiarity with agent frameworks, tool use, or long-horizon task orchestration - Open-source contributions in the ML or AI space

Benefits and work setup

- Fully remote, async-first culture with thoughtful written communication - Direct collaboration with founders and rapid ship-to-production cadence - Equity compensation and high-end hardware provided